BC abaddon
Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious processes, and OpenClaw config posture. Assigns a letter grade. Built for macOS deployments.
Red team security mode for OpenClaw.
As a process C 57/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, running it twice
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
For the model run — optional
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 14
✓ No critical or high findings
Medium and low: 14
-
low Risky intent
intent-offensive-securitySKILL.md:3Offensive-security / dual-use content (legitimate for authorised testing; review intended use)description: Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious processes, and OpenClaw config p
-
low Risky intent
intent-offensive-securitySKILL.md:67Offensive-security / dual-use content (legitimate for authorised testing; review intended use)If you have Gideon (the OpenClaw observer agent), append the red team section:
-
low Risky intent
intent-offensive-securitySKILL.md:98Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)- "run red team"
quoted -
low Risky intent
intent-offensive-securitySPEC.md:1Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)# Abaddon — Red Team Mode (merged into Gideon)
fixture -
low Risky intent
intent-offensive-securitySPEC.md:9Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition; test fixture / example file)Abaddon is not a separate agent. It is Gideon's adversarial mode — an offensive red team scan that runs alongside the standard nightly defensive audit.
detectorfixture -
low Risky intent
intent-offensive-securitySPEC.md:20Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)| Abaddon nightly (offensive) | 3:45 AM CST daily | Adversarial red team | memory/audits/abaddon-YYYY-MM-DD.md |
fixture -
low Risky intent
intent-offensive-securitySPEC.md:21Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)| Abaddon manual | On demand | Adversarial red team | Same, plus letter grade posted to Security topic |
fixture -
low Risky intent
intent-offensive-securitySPEC.md:27Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file; quoted — discussed, not commanded)- "run red team"
fixturequoted -
low Risky intent
intent-offensive-securitytemplates/abaddon-prompt.md:3Offensive-security / dual-use content (legitimate for authorised testing; review intended use)## Red Team Mode (Abaddon) 🔴
-
low Risky intent
intent-offensive-securitytemplates/abaddon-prompt.md:5Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)Activated when triggered with: "run red team", "run Abaddon", "run full assessment", or via nightly cron at 3:45 AM.
quoted -
low Risky intent
intent-offensive-securitytemplates/abaddon-prompt.md:7Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)In red team mode, shift from passive monitor to active attacker. Think like an adversary. Find what they would find.
detector -
low Risky intent
intent-offensive-securitytemplates/abaddon-prompt.md:10Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)If the user says "run red team", "run Abaddon", "run full assessment", or "Abaddon report" — enter this mode immediately. Do NOT run the standard passive audit. Run the full Abaddon protocol below.
detector
A further 2 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when")
Process rating: all ten parameters 57/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 55Failures and branches. 1 branches
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 34 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 848 tokens
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +1No license
- +2Single-language instructions
- +3Description length 247: enough signal without eating the budget
- +4Structure: 10 headings
- +3Step-by-step instructions: 34 items
- +3Output format is stated explicitly
- +4Has examples (4 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.